[00:01] <slangasek> twitch
[00:02] <slangasek> or more importantly, apport :P
[00:12] <SpamapS> hmm...
[00:39] <Keybuk> Completed x11-drivers/xf86-input-cmt-0.0.1-r20 (in 0m3.9s)
[00:39] <Keybuk> argh, ww
[00:43]  * RAOF wonders what -input-cmt applies to.
[00:47] <TheMuso> Y ay for broken java deps.
[00:48] <TheMuso> What do other pilots do if they cannot build a package, fixing the build is non-trivial?
[00:48] <TheMuso> package, and fixing the build is non-trivial even.
[00:49] <TheMuso> I'm looking at visualvm
[00:51] <RAOF> I think it depends.
[00:52] <TheMuso> Well it seems that visualvm is tangled up in lots of java bits, and some or all of that appears to be broken dep wise atm.
[00:53] <RAOF> I know some sponsors will just go ‘eh, it doesn't work, say so on the bug and move on’.
[00:54] <TheMuso> Right, I am just working out whether this is fixable... May be.
[00:59] <TheMuso> Wow netbeans is huge.
[00:59] <lifeless> TheMuso: massive
[01:07] <TheMuso> ...or not. Netbeans FTBFS for another reason, which I wouldn't have a clue on fixing.
[01:26] <maum> hello
[01:26] <maum> can i use touch screen on ubuntu?
[01:27] <lifeless> yes
[01:28] <maum> lifeless, how can I use touch device on ubuntu 11.04? there is no driver for ubuntu.
[01:28] <lifeless> for user suppoer please see the #ubuntu channel. thanks
[01:29] <maum> lifeless, ok
[01:32] <micahg> slangasek: or StevenK, are you either of you around and available to copy something?
[01:41] <slangasek> micahg: hit me
[01:41] <micahg> slangasek: chromium browser from ubuntu-security-proposed to -proposed for lucid-natty
[02:07] <slangasek> micahg: 13.0.782.107~r94237-0ubuntu0.10.04.1 -- is that the right version?
[02:09] <micahg> slangasek: yes
[02:10] <slangasek> micahg: done
[02:10] <micahg> slangasek: thanks!
[02:10] <slangasek> n/p
[02:11] <micahg> slangasek: no, I wanted them to go to proposed :(
[02:20] <micahg> slangasek: ?
[02:25] <slangasek> eh shoot
[02:26] <micahg> slangasek: can you fix or will that break it?
[02:26] <micahg> by break it I mean prevent from publishing later
[02:26] <slangasek> I don't know
[02:27] <slangasek> I can at least try to remove it now, and we can bug LP folks later to manually fix it up for us?
[02:27] <micahg> slangasek: k
[02:29] <maum> I cannot see speaker icon from tray
[02:29] <micahg> maum: support is in #ubuntu
[02:29] <maum> ok
[02:30] <slangasek> micahg: ok, published to proposed, deleted from lucid-security; I don't know if add+delete in a single run will let the previous version stay in place or if it's going to result in it being pulled from -security entirely, but that at least buys us a little time
[02:30] <micahg> slangasek: ok, was just lucid published to -security?
[02:31] <slangasek> micahg: hmm yes, because I had misread 'lucid-natty' as 'lucid-POCKET'
[02:31] <slangasek> so I didn't get a chance to screw up the other two yet
[02:31] <slangasek> let's do that now ;)
[02:31] <micahg> slangasek: heh, ok :)
[02:31]  * micahg gets to testing lucid just in case
[02:32] <TheMuso> @pilot out
[02:51] <micahg> slangasek: thanks, I'll find someone later if there's a problem copying to -security
[03:29] <merlot> hello my ubuntu friends, I have developed a commandline c++ program that fixes a compatibility problem with microsoft mice and X windows. Where should I post this to get included in the ubuntu distribution? Secondly, how should I go about having this run on startup, through /etc/init.d?
[03:33] <merlot> are the mailing lists or forums the best method for such a discussion?
[03:38] <RAOF> merlot: What does this program do?  From your description it sounds like it's working around a bug which would better be fixed elsewhere?
[03:41] <merlot> RAOF: I'm not sure where it fits in. I have developed a linux kernel patch as well, but I'm not sure if it belongs there. There's a couple of Microsoft mice that after being booted into windows, and then rebooted into linux, display a very bad scrolling problem. They scroll 3 or 4 times the amount they should. By using libusb-1.0.8 and wireshark, I was able to find the right codes to send to the mouse to turn off a feature in
[03:41] <merlot>  the mice that cause this problem.
[03:42] <RAOF> Right.  That should be a kernel patch.
[03:43] <RAOF> That indicates that the kernel isn't setting the device up properly; the proper fix would be to make the kernel set the device up properly :)
[03:43] <RAOF> That said, workaround programs can sometimes have their place.
[03:44] <RAOF> But in this case it sounds (to my not-particularly-kernel-developer self) like it should be a fairly simple, and perfectly acceptable, kernel patch.
[03:45] <merlot> I'll ask on the linux kernel mailing list. Thanks for your advice.
[03:47] <merlot> you'll have to excuse me as I wrote this a few months ago and have not had time to post the code anyway. The only mouse I know that it works with is my own model, so I was thinking that it would be better to test this in userland first
[03:47] <merlot> I'll submit the userland libusb code as well.
[03:59] <RAOF> That would be helpful even if your specific patch doesn't get applied.
[04:00] <RAOF> merlot: Have you filed a bug about this, by the way?  That's the best way to ensure that the problem doesn't get forgotten about in the fast-paced world of IRC.
[04:00] <RAOF> Also, the Ubuntu kernel team can help guide you through the process of getting your patch into the kernel.
[04:12] <merlot> there is a bug posted somewhere related to ubuntu or gnome, can't remember which let me dig it back up
[04:15] <maum> https://answers.launchpad.net/ubuntu/+question/167005
[04:24] <merlot> ROAF, I found it here: https://answers.launchpad.net/ubuntu/+source/xorg/+question/9200
[05:15] <didrocks> good morning
[05:42] <merlot> mornin' didrocks
[05:46] <didrocks> hey merlot
[07:58] <hrw> cyphermox: sorry for not appearing later yesterday. got hooked to other meetings
[11:09] <jml> does Dpkg::Shlibs have browsable documentation somewhere?
[11:09] <jml> google is being unhelpful
[11:10] <RAOF> jml: I've always just dived into the source :/
[11:11] <jml> RAOF: I guess I can do that.
[11:11] <jml> ... and overcome my snobbishness against perl
[11:11] <jml> RAOF: also, "always"? I'm kind of surprised that you've needed to.
[11:11] <RAOF> In my experience debhelper perl seems to be reasonably same.
[11:12] <infinity> That would be dpkg-dev, not debhelper.
[11:12] <infinity> Or, should be...
[11:13] <RAOF> Well, yeah.  But same project, so obviously dpkg-dev perl will be sane, too :)
[11:14] <RAOF> jml: I wanted to stop people uploading broken mono packages, so I made dh_clideps fail the build in that case.
[11:15] <infinity> It saddens me that I had to double-check that the world hadn't just gone topsy turvy.
[11:15] <infinity> RAOF: Same project how? :P
[11:16] <RAOF> infinity: Both uploaded to Debian!  That means they're by the same company, right? :)
[11:16]  * infinity blinks.
[11:16] <infinity> If my head explodes all over Terminal 3, it's your fault.
[11:16] <ajmitch> such logic is astounding
[11:17] <RAOF> I reserve the right to make no sense on a Friday evening.
[11:17] <RAOF> And, in fact, at any other time of my, or my body's, chosing!
[11:17] <infinity> Fair enough.
[11:19]  * infinity finds the omission of a manpage for Dpkg::Shlibs a bit odd, now that he looks.
[11:19] <infinity> Almost every other module has one.
[11:26] <jml> actually, Dpkg::Shlibs::SymbolFile was what I really wanted.
[11:26] <jml> that probably has no man page on account of having no documentation in the code either.
[11:28] <tkamppeter> I want to merge two source packages into one, as they come from the same upstream tarball: argyll and libicc2.
[11:28] <tkamppeter> The resulting binary packages will stay the same: libicc, libicc-dev, argyll, and argyll-dbg.
[11:29] <tkamppeter> Problem is that the upstream tarball version is 1.3.3, but libicc is 2.12. How do I do the merge then so that updates work.
[11:31] <geser> tkamppeter: the new libicc version is also 1.3.3? or does libicc still have its own upstream version?
[11:33] <tkamppeter> geser, I do not know wehre the 2.12 comes from, probably it is the API version (libicc.so.2.12), which is wrong. A package must have the upstream version and as both come from the same tarball, the upstream version is 1.3.3.
[11:34] <tkamppeter> Yes, the 2.12 is the API vewrsion. The lib file is /usr/lib/libicc.so.2.12.0.
[11:36] <geser> but that would only affect the package naming (I'm not that firm in library packaging)
[11:36] <geser> but if upstream uses now a different versioning scheme and you need to go backwards add an epoch
[11:37] <tkamppeter> geser, what is the best to do here, especially important also is that Debian will want to merge it back.
[11:39] <tkamppeter> geser, Upstreame has probably never advanced to 2.12 with their version number for Argyll. The 2.12 is the API version and in package naming it usually only reflected by saying "libicc2" instead of "libicc".
[11:42] <geser> yes, I guess using libicc2 would be a good idea (perhaps even libicc2-12; but better ask someone with more knowledge in library packaging about this) and add an epoch for the version
[11:43] <geser> it would be also a good idea to coordinate this merging with the Debian maintainer to lessen the pain when merging Debian->Ubuntu
[12:11] <ScottK> bdmurray: Rather than me get in a reversion war, would you please look at Bug #745836 and see if you think listing "Use Gnumeric instead" as a workaround for the bug is a good idea.
[12:15] <mvo> brendand: hello! I believe your update-manager branch about the battery-msg is no longer needed, could you mark it as obsoleted please? the gtk3 port fixes it for good :)
[12:30] <brendand> mvo - sure
[12:49] <ahasenack> SpamapS: hi, any news on the landscape-client sru?
[13:51] <ryanakca> Is there a list of source packages that have been removed from the archive?
[14:01] <Laney> ryanakca: using the LP API, yes
[14:01] <Laney> don't know if there is another way
[14:05] <ryanakca> Laney: Would those be all the sources with the "Deleted" status?
[14:05] <Laney> seems so
[14:06] <Laney> you might want to restrict the pocket and/or distro_series, depending on what you want
[14:06] <Laney> (all the souce_package_publishing_history records, that is)
[14:08] <ryanakca> Thanks
[15:07] <hallyn> i keep finding UDD trees out of date.  So I fetch the pkg from archive, bzr import-dsc, then push.  Is that wrong?
[15:07] <micahg> hallyn: probably better to get the importer fixed, it'll probably fail again next time
[15:09] <hallyn> micahg: http://package-import.ubuntu.com/status/qemu-kvm.html#2011-05-16%2017:12:45.463425    If I read that right, it's complaining about changes to debian/qemu-kvm.1.  If that's the case, then manually pushing it should solve that right?
[15:09] <hallyn> i.e., next time it shouldn't fail again?
[15:14] <micahg> hallyn: probably, but you still might want to check to see that they have a bug for that situation if it's indeed a bug
[15:18] <hallyn> micahg: thanks, i'll file a bug
[15:34] <SpamapS> ahasenack: landscape-client was released to lucid-updates yesterday
[15:34] <ahasenack> SpamapS: thanks! I didn't see the LP email flood that normally happens, so I was wondering
[15:35] <SpamapS> ahasenack: landscape-client | 11.07.1.1-0ubuntu0.10.04.0 | lucid-updates | source, amd64, i386
[15:35] <ahasenack> SpamapS: \o/
[15:36] <micahg> SpamapS: how can you do that?  doesn't that produce regressions on upgrade? (maverick, natty not in -proposed)
[15:36] <SpamapS> micahg: we do lucid only SRU's all the time.
[15:37] <SpamapS> micahg: certainly not the best way to go about things
[15:37] <micahg> yes, but aren't those usually for stuff only found in Lucid?
[15:37] <micahg> after maverick EOL it's certainly fine, but I would think while there's an upgrade path, regression-release issues should be minimized where possible
[15:39] <SpamapS> As I said, certainly not the best way to go about things, but actually they'll just keep the lucid version on upgrade.
[15:40] <SpamapS> I believe 11.07 is also supposed to arrive in -proposed soon enough.
[15:40] <micahg> right which can potentially break other things or itself depending on what the package is
[15:45] <mdeslaur> slangasek: mind if I merge libpng? (you TIL)
[15:47] <SpamapS> ahasenack: you guys do plan to push 11.07 into maverick and natty, right?
[15:48] <ahasenack> SpamapS: right, once those remaining tests are completed, hopefully sometime next week
[15:49] <SpamapS> ahasenack: what additional tests? landscape-client is not in natty/maverick proposed
[15:50] <ahasenack> SpamapS: right, I will only ok it once the pre-upload tests are done
[15:50] <SpamapS> ah ok
[15:50] <ahasenack> SpamapS: we test it throughly before allowing it to be uploaded to -proposed
[15:50] <ahasenack> thoroughly
[15:50] <ahasenack> SpamapS: the test matrix url is in the bug, if you are curious
[15:51] <SpamapS> ahasenack: to be clear, I think I may have actually flubbed up a bit by letting it into lucid-updates before it was available for all other releases.. as michag points out.. it will cause issues for people on upgrade. So maybe next time upload them all at once, or work backward, from natty -> lucid
[15:51] <ahasenack> SpamapS: hmm, really? What issues?
[15:52] <ahasenack> SpamapS: like lucid people upgrading to maverick, you mean? The package won't be upgraded?
[15:52] <SpamapS> ahasenack: right
[15:53] <ahasenack> SpamapS: ok, I hand't thought of that, I'll speed up the testing
[15:53] <hrw> cyphermox: may we talk today?
[15:53] <cyphermox> hrw:  certainly
[15:53] <hrw> cyphermox: cool
[15:54] <hrw> cyphermox: so how can I help with this nm-applet bug?
[15:54] <cyphermox> which?
[15:54] <hrw> bug 819700
[15:55] <SpamapS> ahasenack: ty
[15:56] <cyphermox> hrw: libappindicator fallbacks appear to be broken in that they don't properly update the menu when that gets called by nm-applet; might be an issue with the dbusmenu about-to-show signal
[15:56] <cyphermox> hrw: essentially this need to be debugged at the libappindicator level I think
[15:58] <hrw> cyphermox: any hints on that?
[15:59] <hrw> cause now I feel that only unitypanel has working appindicator
[15:59] <cyphermox> well, I'll need to dig into the fallback code, it's clearly marked as that
[16:00] <cyphermox> hrw: no, xubuntu has it working too
[16:00] <hrw> cyphermox: thats xubuntu/oneiric
[16:00] <hrw> cyphermox: under natty it was working, broke on oneiric
[16:00] <cyphermox> it depends whether whatever panel in use has indicator widgets thingy
[16:00] <cyphermox> yeah
[16:01] <cyphermox> I changed nm-applet so that it would update the menu when you click the icon rather than only when e.g. a new ap is discovered
[16:02] <hrw> cyphermox: xfce4 indicator-plugin for panel just says 'No indicators'
[16:03] <hrw> and I have nm and bt running iconless
[16:03] <cyphermox> something ought to be broken there, because on my newly installed system it works and displays nm-applet and bluetooth and all
[16:03] <hrw> (xfce4-indicator-plugin:5128): GLib-GObject-WARNING **: invalid uninstantiatable type `(null)' in cast to `GtkWidget'
[16:03] <hrw> (xfce4-indicator-plugin:5128): Gtk-CRITICAL **: IA__gtk_widget_set_name: assertion `GTK_IS_WIDGET (widget)' failed
[16:03] <hrw> (xfce4-indicator-plugin:5128): libxfce4panel-CRITICAL **: IA__xfce_panel_plugin_add_action_widget: assertion `GTK_IS_WIDGET (widget)' failed
[16:05] <hrw> http://pastebin.com/Gp9F7TQM is what I got when started nm-applet
[16:54] <micahg> cyphermox: on oneiric?
[17:02] <SpamapS> Hmm.. when writing an apport hook.. how can I get access to what will eventually be the bugs title?
[17:04] <nigelb> bdmurray would know
[17:06] <SpamapS> actually I think I found a better way which will just add the info all the time as its useful in other contexts.. :)
[17:31] <SpamapS> bdmurray: whats the best way to test an apport package hook?
[17:31] <SpamapS> bdmurray: break the postinst and dpkg-reconfigure the package?
[17:32] <nigelb> SpamapS: wait, why?
[17:32] <nigelb> SpamapS: testing is easy.
[17:32] <SpamapS> I want to test its handling of postinst failures
[17:33] <nigelb> oh. that. :)
[17:33] <nigelb> probably :)
[18:00] <cyphermox> micahg: yes, I'm on oneiric
[18:02] <SpamapS> lamont: were you aware that Postfix FTBFS in oneiric?
[18:02] <SpamapS> http://tech.groups.yahoo.com/group/postfix-users/message/278015
[18:15] <micahg> cyphermox: is this a live CD?
[18:15] <cyphermox> micahg: no?
[18:15] <cyphermox> micahg: I can try things on a live cd though
[18:16] <cyphermox> got xubuntu desktop amd64 already retrieved
[18:16] <micahg> cyphermox: weird, the network manager applet works fine for me on oneiric
[18:16] <cyphermox> micahg: yes
[18:16] <cyphermox> micahg: the problem is for those who don't have the indicator widget thing for xfce or for other wms
[18:17] <micahg> cyphermox: indicator widget?
[18:17] <micahg> oh, I should be asking hrw :)
[18:17] <cyphermox> well, whatever thing that will allow indicators to be added to a panel in the specific WM being using
[18:17] <micahg> well, we have an indicator plugin that allows us to use the indicators with the xfce4-panel
[18:18] <cyphermox> yes
[18:18] <cyphermox> but I think one of the two available xubuntu sessions will use the old systray instead no?
[18:18] <cyphermox> (and there may be other things at play here, like people actually removing some of these bits for whatever reason)
[18:19] <charlie-tca> no
[18:19] <charlie-tca> one of the two will be a cross between the xfce and xubuntu sessions, with most of the stuff in the panels missing
[18:19] <cyphermox> in any case, nm-applet should work properly for those who can't use indicators; the fallback is clearly broken in many ways, including displaying icons for any such indicators, not just nm-applet
[18:20] <cyphermox> charlie-tca: ok, thanks for the clarification
[18:20] <micahg> well, the xfce4-indicator-plugin is removable as it's a recommends of xubuntu-desktop
[18:20] <cyphermox> charlie-tca: but in one of those wouldn't you see stuff in the panel showing broken icons?
[18:20] <charlie-tca> and it takes 12 steps listed in the alpha3 technical overview to get the right session if you pick the wrong one. too
[18:20] <cyphermox> fun
[18:21] <charlie-tca> yes, but I can't tell if it is because it is missing, or because it picked things xfce session should not have
[18:21] <cyphermox> bah, all the icons should be there for nm-applet and bluetooth. they usually don't change
[18:22] <cyphermox> if it's not showing for just those two it's definitely an issue in libappindicators, as those are the two major "applets" using that library
[18:22] <charlie-tca> It is the things like indicators not working/missing, 4 of eight launchers missing, 4 workspaces instead of two, network manager applet doesn't show the connections
[18:22] <cyphermox> charlie-tca: yes
[18:22] <charlie-tca> which is a mixed session, it is not xfce or xubuntu
[18:23] <cyphermox> ok
[18:23] <charlie-tca> but if it helps... just delete the ~/.config and pick the other session to login
[18:24] <cyphermox> anyway, nm-applet not showing the connections is a side effect of (xfce4-indicator-plugin missing || something broken in libappindicator)
[19:28] <ScottK> SpamapS: lamont is on vacation this week.  The patch you pointed to in the bug is the one we want, so please feel free to fix it.
[19:43] <jono> hi folks
[19:43] <jono> is it me, or are there a stack of system settings missing?
[19:43] <jono> things like the screensaver config tool
[19:43] <jono> fonts
[19:43] <jono> managing how often the screen dims etc
[19:44] <kirkland> mvo: what would you think of adding a flag to apt-add-repository that triggers an immediate apt-get update after a successful PPA add?
[19:44] <kirkland> mvo: it just seems I *always* end up apt-add-repository ppa:foo/ppa && apt-get update
[19:45] <kirkland> jono: they've kinda moved around a little, i think
[19:45] <jono> kirkland, that's what I thought, but big pieces are missing
[19:45] <jono> for example, I can't find the screensaver and screen dimming options
[19:45] <cyphermox> jono: screensaver: indeed, you can't choose your screensaver, only the settings for when to lock, etc; that's also where you set dimming
[19:45] <jono> cyphermox, wow, that seems like quite a regression
[19:45] <cyphermox> under the power menu thing, clikc System Settings, and under Personal you should have something called "screen"
[19:46] <cyphermox> jono: reimplementing screensavers was discussed for oneiric but we haven't had time afaik
[19:46] <kirkland> jono: alt-f2, and then "system settings"
[19:46] <kirkland> jono: that's the best I can find, but you're right, there are config settings I can't find any more
[19:46] <jono> cyphermox, so this was just dropped from the GNOME3 Control Center re-write?
[19:46] <cyphermox> jono: sadly, yes
[19:47] <jono> wow
[19:47] <jono> thanks for the clarification
[19:47] <kirkland> jono: i want to disable suspend when i close my laptop lid (I friggin hate that setting)
[19:47] <cyphermox> as far as fonts go, afaict the only way is to install gnome-tweak-tool
[19:47] <jono> I am amazed GNOME3 shipped like this
[19:47] <jono> seems like a lot of things are missing
[19:47] <cyphermox> (and gnome-tweak-tool... urkk)
[19:48] <ScottK> cyphermox: Switching to KDE is another option. ;-)
[19:48] <jono> :-)
[19:48] <cyphermox> ScottK: indeed ;)
[19:48] <cyphermox> ScottK: how's the NM plasmoid ;D
[19:48] <ScottK> cyphermox: Seems to be working.
[19:49] <ScottK> I only had a spare computer to run Oneiric with starting this week, but it's worked fine so far.
[19:49] <cyphermox> ok
[19:49] <cyphermox> I wonder if it's still far out of date from the NM api, if it is, there is still time to fix that
[19:50] <ScottK> cyphermox: There's a NM09 branch that we are using.
[19:50] <cyphermox> ScottK: ah, that's what I was wondering
[19:50] <cyphermox> perhaps it's time I give kubuntu a honest shot again
[19:50]  * charlie-tca thinks about Linus saying Xfce
[19:51]  * charlie-tca knows there is a Xubuntu with Xfce, too
[19:51] <Laney> a lot of that gnome 3 stuff is supposed to be coming back
[20:07] <slangasek> mdeslaur: not at all
[20:11] <mvo> kirkland: good idea
[20:12] <kirkland> mvo: thanks!
[20:12] <kirkland> mvo: you want a bug or do you want to JFDI?  :-)
[20:15] <vmlinuz> anyone familiar with networkmanager-vpnc build ?
[20:16] <vmlinuz> on Natty, pkg-config reports NetworkManager is 0.8.3.998
[20:17] <vmlinuz> if you try to build network-manager-vpnc directly from svn, it fails to recognize NetworkManager and its libs
[20:17] <vmlinuz> Requested 'NetworkManager >= 0.8.998' but version of NetworkManager is 0.8.3.998
[20:17] <mvo> kirkland: a small reminder maybe on monday or a bugreport please, otherwise I will have forgotten it over the weekend, I know myself
[20:17] <kirkland> mvo: bug report it is
[20:17] <mvo> thanks!
[20:17]  * mvo calls it a day
[20:18] <vmlinuz> /usr/lib/pkgconfig/NetworkManager.pc has "Version: 0.8.3.998"
[20:19] <vmlinuz> but networkmanager package is version 0.8.4~git.20110319t175609.d14809b-0ubuntu3
[20:20] <vmlinuz> I'd say the bug is on the "network-manager-dev" package, right?
[21:14] <dupondje> http://reports.qa.ubuntu.com/reports/sponsoring/
[21:14] <dupondje> something broke ? :P
[22:45] <SpamapS> ScottK: thanks. So the patch should just be in the tree right, no patchsys ?
[22:48] <ScottK> Yep.  lamont puts everything in a full git tree.